A hybrid approach for selective sulfoxidation via bioelectrochemically

Subscriber access provided by Kaohsiung Medical University

A hybrid approach for selective sulfoxidation via bioelectrochemically derived hydrogen peroxide over a niobium(V)-silica catalyst James S. Griffin, Eric Taw, Abha Gosavi, Nicholas E Thornburg, Ihsan Pramanda, Hyung-Sool Lee, Kimberly A Gray, Justin M Notestein, and George F. Wells ACS Sustainable Chem. Eng., Just Accepted Manuscript • DOI: 10.1021/ acssuschemeng.7b04641 • Publication Date (Web): 27 Apr 2018 Downloaded from http://pubs.acs.org on April 27, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

ACS Sustainable Chemistry & Engineering

A hybrid approach for selective sulfoxidation via bioelectrochemically derived hydrogen peroxide over a niobium(V)-silica catalyst James Griffin1, Eric Taw1, Abha Gosavi1, Nicholas E. Thornburg1†, Ihsan Pramanda2, HyungSool Lee3, Kimberly A Gray4, Justin M. Notestein1, George Wells4* 1

Department of Chemical and Biological Engineering, Northwestern University, 2145 Sheridan Rd. E-136, Evanston, IL 60208. 2

Master of Science in Biotechnology, Northwestern University, 2145 Sheridan Rd. E-136, Evanston, IL 60208. 3

Department of Civil and Environmental Engineering, University of Waterloo, 200 University Avenue West, Waterloo, ON Canada N2L 3G1.

4

Department of Civil and Environmental Engineering, Northwestern University, 2145 Sheridan Rd. A-318, Evanston, IL 60208.

Keywords: microbial peroxide producing cell, bioelectrochemical system, green chemistry, catalysis, sulfoxidation, resource recovery

25 26

*Corresponding author: George F. Wells, A318 Technological Institute 2145 Sheridan Road

27

Evanston, IL 60208-3109, Phone: (847) 491-9902, Fax: (847) 491-4011, Email:

28

[email protected]

29

Submission as a Research Article to ACS Sustainable Chemistry and Engineering

30 31

ACS Paragon Plus Environment

ACS Sustainable Chemistry & Engineering 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

32

Abstract:

33

In this work, we demonstrate a combined bioelectrochemical and inorganic catalytic

34

system for resource recovery from wastewater. We designed a microbial peroxide producing cell

35

(MPPC) for hydrogen peroxide (H2O2) production and used this bioelectrochemically-derived

36

H2O2 as a green oxidant for sulfoxidation, an industrial reaction used for chemical synthesis and

37

oxidative desulfurization of transportation fuels. We operated an MPPC equipped with a gas

38

diffusion electrode cathode for six months, achieving a peak current density above 1.4 mA cm-2

39

with 60% average acetate removal and 61% average anodic coulombic efficiency. We evaluated

40

several cathode buffers under batch and continuous flow conditions for solubility and pH

41

compatibility with downstream catalytic systems. During 24-hour batch tests, a phosphate

42

buffered MPPC achieved a maximum H2O2 concentration of 4.6 g L-1 and a citric acid-phosphate

43

buffered MPPC obtained a moderate H2O2 concentration (3.1 g L-1) at a low energy input (1.6

44

Wh g-1 H2O2) and pH (10). The MPPC-derived H2O2 was used directly as an oxidant for the

45

catalytic sulfoxidation of 4-hydroxythioanisole over a solid niobium(V)-silica catalyst. We

46

achieved 82% conversion of 50 mM 4-hydroxythioanisole to 4-(methylsulfinyl)-phenol with

47

99% selectivity with a 0.5 mol% catalyst loading in 100 minutes in aqueous media. Our results

48

demonstrate a new and versatile approach for valorization of wastewater through continuous

49

production of H2O2 and its subsequent use as a selective green oxidant in aqueous conditions for

50

green chemistry applications.

51


Page 2 of 30

Page 3 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


52



53

Introduction:

54

Modern wastewater treatment is vital for protecting human and environmental health.

55

However, most design decisions are based on a cost benefit analysis that does not fully value

56

resource recovery. This has resulted in an unsustainable “once-through” model that focuses on

57

waste removal and disposal. Bioelectrochemical systems (BESs) use a biological catalyst to

58

produce electrical current from wastewater.1 Most BESs have focused on producing electricity in

59

microbial fuel cell (MFC) configurations; however, there are some examples of cathodic

60

electrochemical synthesis including hydrogen (H2) or hydrogen peroxide (H2O2) production in

61

microbial electrolysis cells (MECs) or microbial peroxide producing cells (MPPCs),

62

respectively.2-3 A life cycle assessment of MPPCs found that at scale, MPPC-derived H2O2 has

63

nearly 60% lower life cycle greenhouse gas emissions than H2O2 produced via the traditional

64

anthraquinone auto-oxidation process.4 Previous studies on H2O2 production in MPPCs5-7 have

65

focused on producing alkaline H2O2 for disinfection, industrial bleaching or non-specific

66

advanced oxidative processes such as the bio-electro-Fenton process for removal of recalcitrant

67

contaminants from water8.

68

In addition to its proven uses in disinfection and bleaching, H2O2 is a promising green

69

oxidant for sustainable chemistry applications because, unlike organic hydroperoxides, it

70

generates water as its only byproduct.9-11 While the use of H2O2 with solid oxide catalysts to

71

selectively oxidize sulfides has been extensively studied,12-15 there are few examples of selective

72

oxidation in these systems in aqueous conditions, and none involving bioelectrochemically

73

derived peroxide.16 We recently demonstrated high rates and selectivities in the oxidation of

74

thioanisole and several benzothiophene derivatives by H2O2 over a highly dispersed supported

75

niobium(V)-silica (Nb(V)-SiO2) catalyst in acetonitrile, which outperformed benchmark titania-

76

silica and zirconia-silica materials.17 Thus, sulfoxidation of a thioanisole derivative over a

77

similarly formulated niobium(V)-silica catalyst was chosen as a model reaction system for proof-

78

of-concept testing of the direct use of bioelectrochemically derived H2O2 for selective oxidation

79

in an aqueous buffer (rather than in acetonitrile or other organic solvents). This approach could

80

be adapted to the epoxidation and/or dihydroxylation of alkenes,18 the oxidative

81

depolymerization of lignin macromolecules19, or other H2O2-dependent catalytic processes.

82

H2O2 can be generated for these applications via electrocatalytic oxygen reduction. In

83

BESs, anode respiring bacteria anaerobically consume organic matter using an external electrode


Page 4 of 30

Page 5 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


84

as a terminal electron acceptor via a metabolic process known as extracellular electron transfer

85

(EET).20 Current produced during EET can reduce oxygen to water via a four-electron reduction

86

(Equation 1). Oxygen can also be partially reduced to H2O2 via a two-electron oxygen reduction

87

reaction that occurs on graphitic cathodes (Equation 2).21 H2O2 can be reduced further, resulting

88

in a net production of water and decreased coulombic efficiency in an MPPC.22

89

+ 4 + 4 → 2 = 0.81V23

(Eq. 1)

90

+ 2 + 2 → = 0.28V23

(Eq. 2)

91

H2O2 synthesis in acetate-fed MPPCs is exergonic,7 although most previous efforts have applied

92

external voltage to improve H2O2 synthesis rates and titers at the cost of increasing energy

93

intensity. Rozendal et al. developed the first bioelectrochemical H2O2 reactor and achieved

94

concentrations of 1.9 g L-1 at an efficiency of 83% and energy input of 0.93 Wh g-1 H2O2 using

95

an acetate media as feed.5 Fu et al. demonstrated that H2O2 could be generated with a net positive

96

energy production but achieved a maximum concentration of 79 mg L-1.7

97

System design improvements such as minimizing electrode spacing3 and the use of

98

composite carbon black-graphite-PTFE cathodes24-26 have led to higher efficiency and lower

99

system overpotentials. Despite recent improvements in system performance and cell design, “pH

100

splitting”, or opposing pH shifts at the anode and cathode chambers, remains a major problem

101

limiting MPPC H2O2 titer. Proton consumption during oxygen reduction raises catholyte pH and

102

commonly used ion exchange membranes primarily transport ions other than hydroxide or

103

protons at typical ionic strengths.27 Elevated cathode pH decreases coulombic efficiency by

104

promoting H2O2 decomposition28 and increases cell overpotential in MPPCs, thus increasing

105

energy input per gram of H2O2 produced. In a recent study, 80% of the initial H2O2 stored in pH

106

12.5 NaCl electrolyte solution decomposed in three days.3 For the latter reason, there is

107

significant motivation for the immediate use of the produced H2O2 in a continuous process.

108

In this study, we present a novel hybrid biological and chemical catalytic approach to

109

wastewater resource recovery via bioelectrochemical H2O2 production and subsequent use of the

110

H2O2 solution as a green oxidant in chemical synthesis via heterogeneous catalysis. Specifically,

111

we demonstrate thioether sulfoxidation, an industrially relevant process for the synthesis of

112

medicinally relevant sulfoxides and sulfones29 and for the removal of sulfur compounds from

113

fuels and industrial effluents. First, we optimized MPPC cell design, hydraulic residence time

114

and cathode buffer composition to maximize H2O2 concentration at a pH compatible with a



115

silica-based catalyst. We operated an MPPC using a continuous flow cathode for six months and

116

produced H2O2 at average effluent concentrations of 3.2 g/L at a 40-hour HRT. Next we

117

characterized the kinetics and selectivity of catalytic sulfoxidation of 4-hydroxythioanisole to 4-

118

(methylsulfinyl)-phenol by H2O2 in aqueous buffer solutions and used that to design and operate

119

a system for continuous sulfoxidation using the MPPC effluent.

120 121

Methods:

122 123

MPPC Reactor Configuration:

124

Bioelectrochemical cells were constructed from laser-cut acrylic with an anode chamber

125

volume of 200 mL equipped with an Ag/AgCl reference electrode (BASI, West Lafayette, IN).

126

During MPPC operation, a 25 cm2 piece of AvCarb carbon felt (Fuel Cell Earth, Woburn, MA)

127

with a copper mesh current collector was used as the anode. The carbon felt was pretreated for

128

24 hours in 1N nitric acid, followed by 24 hours in acetone and 24 hours in ethanol.30 During

129

abiotic cathode optimization, the copper-backed carbon felt was replaced by a 10 cm2 Pt mesh

130

(Sigma) that was used as a counter anode. The cathode chamber had a 100 mL chamber with an

131

exposed cathode surface area of 25 cm2. A gas diffusion cathode was constructed as previously

132

described24 using a carbon cloth electrode (GDL-CT, Fuel Cells Etc., College Station, TX)

133

coated with PTFE on one side and 5 mg cm-2 carbon black (Vulcan XC-72, Fuel Cell Earth,

134

Woburn, MA) dispersed in Nafion on the other side. The chambers were separated by a 33 cm2

135

anion exchange membrane (Ultrex AMI-7001, Membranes Intl., Ringwood, NJ). The electrode

136

spacing was 3.4 cm. Both chambers contained inlet and outlet ports to operate in continuous

137

mode. During continuous H2O2 production and usage, cathode effluent was stored in a holding

138

tank and neutralized and diluted before being used for sulfoxidation. A schematic of the

139

combined MPPC and sulfoxidation reactor setup is shown in Figure 1, and photographs of the

140

reactor setups are available in Figure S1.

141 142

MPPC Operating Conditions

143

The anodic biofilms used in this study were inoculated with 80 vol% synthetic acetate

144

media, 10 vol% primary effluent from the O’Brien Water Reclamation Plant (Skokie, IL), and 10

145

vol% effluent from a previously operated MPPC. The anode potential was set at -0.3V vs.


Page 6 of 30

Page 7 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


146

Ag/AgCl using a VMP-3 potentiostat (Bio-logic USA, Knoxville, TN) and cell current, working

147

and counter electrode potentials were monitored every 30 seconds. The anode was fed with a

148

synthetic wastewater medium containing 25 mM CH3COONa in a 100 mM phosphate buffer31.

149

Each liter of anode medium also contained 0.3 mL of a trace metal solution.32 MPPC anodic

150

hydraulic residence time (HRT) was initially 12 hours and then decreased to 5 hours as acetate

151

consumption rates increased to avoid substrate limitation.33 The medium was continuously

152

sparged with N2 gas to minimize oxygen introduction into the system. The reactor was operated

153

continuously for over six months, during which time cathode HRT and buffer concentration were

154

varied to study the effect of these variables on efficiency and effluent H2O2 concentration. All

155

MPPC experiments were performed at room temperature (22 ±1°C) with continuous mixing.

156

To identify buffer systems compatible with downstream catalytic processes, H2O2

157

production was tested in abiotic cells equipped with Pt anodes. H2O2 production rates, cathodic

158

coulombic efficiency and catholyte pH increase were evaluated in 24 hour batch assays with

159

several phosphate and citric acid buffer compositions, listed in Table S1. Cathode potential was

160

fixed at -0.6V vs. Ag/AgCl and samples were collected for pH and H2O2 measurement. During

161

continuous MPPC operation, catholyte chambers were triple rinsed with deionized water and

162

then allowed to accumulate H2O2 for 24 hours before starting continuous flow. Effluent H2O2

163

concentrations were monitored until they reached steady state, and effluent was collected for

164

sulfoxidation.

165 166

Analytical Methods:

167

A Trace 1310 gas chromatograph (GC) (Thermo Scientific, USA) equipped with a

168

diphenyl dimethyl polysiloxane column and an FID detector was used to measure anolyte acetate

169

concentrations. A 0.1 µL sample was injected in the GC with a 50:1 split ratio under a constant

170

flow rate of 1.0 ml min-1. The GC oven was held at 70 °C for one minute, then increased at 10 °C

171

per minute up to 180 °C, and then held at a constant temperature of 180°C for five minutes. 24

172

hour anodic acetate removal rates were calculated by measuring the difference between acetate

173

medium and effluent concentrations at a given HRT. H2O2 concentration was measured in the

174

catholyte effluent using the ammonium metavanadate method.34 A vanadate colorimetric reagent

175

was prepared by slowly adding 9 M sulfuric acid to a 12.4 mM ammonium metavanadate

176

solution under magnetic stirring at 50°C until complete dissolution. Triplicate 0.75 mL catholyte



Page 8 of 30

177

samples were diluted to 20-200 mg L-1 H2O2 in deionized water and added to an equal volume of

178

the vanadate solution. Absorbance was measured at 450 nm on a BioSpectrometer UV-Vis

179

spectrophotometer (Eppendorf). Anodic and cathodic coulombic efficiencies were calculated

180

based on the total current and measured changes in acetate and H2O2 concentrations using

181

equations 3 and 4.23

182

Anode coulombic efficiency (%):

183

Cathode coulombic efficiency (%):

184

where I is the total current, n is the electron equivalents per mole of H2O2 or acetate (2 and 8,

185

respectively), [CH3COO-] and [H2O2] are the concentrations of acetate and H2O2, respectively, in

186

mol L-1, V is the anode volume and Q is the volumetric flow rate of catholyte. Overall coulombic

187

efficiency was calculated by multiplying the anodic and cathodic coulombic efficiencies

188

together. Aerial current and power density were calculated by normalizing current and power to

189

the anode and cathode surface area (25 cm2). Batch cathodic coulombic calculations were

190

performed using the volume of catholyte remaining in the reactor while sampling to account for

191

changes in catholyte volume during the batch assays.

∆[ ]"#$%&' [( ( ])

(Eq. 3) (Eq. 4)

192 193

Catalyst Synthesis and Characterization:

194

The Nb–SiO2 catalyst was prepared by grafting a niobium(V)–calixarene coordination

195

complex to partially-dehydroxylated silica gel (Alfa Aesar, 75-150 µm, 373 m2 g-1 BET)

196

following similar procedures recently reported by some of us and described in the Supporting

197

Methods.35-36 Catalyst metal content was quantified using inductively coupled plasma atomic

198

emission spectroscopy (ICP-AES; Thermo iCAP 7600). Prepared Nb(V)-SiO2 catalysts were

199

digested in 48 wt % HF, diluted with 0.9 wt % HNO3 and calibrated against a standard curve of a

200

commercial Nb solution (Fluka Analytical) in 0.9 wt % HNO3. Diffuse reflectance UV-visible

201

(DRUV-vis) spectra were collected from 800–200 nm at ambient conditions using a Shimadzu

202

UV-3600

203

polytetrafluoroethylene (PTFE) powder as a baseline perfect reflector and a 20:1 catalyst diluent.

with

a

Harrick

Praying

Mantis

diffuse

reflectance

accessory,

using

204 205 206

4-hydroxythioanisole Oxidation: 4-hydroxythioanisole oxidation kinetics were first investigated across a range of


Page 9 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


207

temperatures, catalyst loadings and aqueous buffer conditions using commercial H2O2 in a

208

“mock MPPC catholyte” as follows: 1 mmol 4-hydroxythioanisole (Sigma) and 0.1 mmol phenol

209

(internal standard, Sigma) were dissolved in a 2:1 deionized water: ethanol solution. Mock

210

MPPC catholyte was produced by pH adjusting a 200 mM pH 6 phosphate buffer (21 g L-1

211

KH2PO4 (Sigma) and 6.6 g L-1 Na2HPO4 . (H2O)7 (Sigma)) to pH 13 with NaOH and then

212

neutralized with HCl to pH 7 to mimic the neutralization procedure used for the MPPC catholyte

213

and achieve a similar chloride concentration in the mock and actual MPPC catholyte. 31.1 mg

214

Nb(V)-SiO2 (0.16 mmol Nb g-1) was then added to a solution containing 15 mL of the 4-

215

hydroxythioanisole solution and 5 mL of mock MPPC effluent. Finally, 110 µL of H2O2 (30 wt

216

% aqueous, 1.07 mmol) was added to start the reaction. Experiments were run between 25°C and

217

75°C on a hotplate and magnetically stirred at 500 rpm. At the start of reaction, the 4-

218

hydroxythioanisole concentration was 50 mM and the molar ratios inside of the reactor were

219

100:107:0.5 for 4-hydroxythioanisole:H2O2:Nb. Reaction aliquots (approx. 200 µL) were

220

quenched with approximately 5 mg NaHSO3 (Sigma) at desired time points before GC analysis.

221

Product identities were initially determined using GC-MS (Shimadzu QP2010, Zebron ZB-624

222

capillary column) and quantified using GC-FID (Shimadzu 2010, TR-1 capillary column) via

223

calibrated standards. The reaction was assumed to be first order in both H2O2 and 4-

224

hydroxythioanisole, giving second order overall, and rate constants were found by a satisfactory

225

linear regression of ln

226 89

,-. ,/0-. 1

= 23 /4 − 1167

(Eq. 5)

227

where 4 =

228

H2O2, respectively. Apparent activation barriers were calculated from the slope of ln(k) vs.

229

1000/T (K) conforming to an Arrhenius model.

.9

= 1.1, k is the apparent rate constant, and A and B as 4-hydroxythioanisole and

230

4-hydroxythioanisole oxidation kinetics were quantified with MPPC-derived H2O2 using

231

a similar procedure. The reactor sampling and GC quantification were as described above, but

232

the mock MPPC catholyte and 30 wt% H2O2 were replaced with H2O2-laden MPPC effluent

233

adjusted to pH 7 with HCl. The volumes of MPPC effluent and deionized water added to the 4-

234

hydroxythioanisole solution were adjusted to give initial molar ratios of 100:110:0.5 for 4-

235

hydroxythioanisole:H2O2:Nb in the reactor.

236

In addition to batch experiments, continuous stirred tank reactor (CSTR) sulfoxidation



Page 10 of 30

237

experiments were performed in a 50 mL stirred round bottom flask controlled at 35 ºC with a 20

238

mL holdup volume. Separate MPPC effluent and 4-hydroxythioanisole solutions were pumped

239

into the reactor at 5 mL hr-1 each using a Minipuls 3 multichannel pump (Gilson, Inc., Middleton

240

WI) to give a two-hour HRT. H2O2 was generated in an MPPC operated at sufficiently long

241

HRTs (on average 1.7 days) in order to reach 110 mM H2O2 (3.7 g H2O2 L-1). MPPC effluent

242

was acidified with HCl to pH 7 prior to usage. The sulfide reactant solution initially contained

243

100 mM 4-hydroxythioanisole and 40 mM phenol dissolved in a 50:50 solution of water and

244

ethanol. The reactor was initially filled with 10 mLs of 4-hydroxythioanisole and catholyte

245

solutions each and contained 0.5 mol% Nb(V)-SiO2 (relative to 4-hydroxythioanisole). Cotton

246

dead-end filters were used to minimize catalyst loss through the effluent ports on the CSTR.

247

Effluent sulfide product distribution (unreacted sulfide [4-hydroxythioanisole], product sulfoxide

248

[4-(methylsulfinyl)-phenol], and sulfone [4-(methylsulfonyl)-phenol] concentrations) was

249

measured as in the batch experiments, and the reactor was operated until the product distribution

250

reached steady state. The apparent rate constant was determined from the effluent concentration

251

from the CSTR using the Levenberg-Marquardt nonlinear least squares regression algorithm and

252

the Runge-Kutta method for numerically integrating differential equations in MATLAB R2016b

253

(The Mathworks, Natick MA). The nlparci function was used to obtain a 95% confidence

254

interval. The effluent concentration was fit to the following differential equation, derived from a

255

mass balance:

256 257

.

=

.9 . :

− 623 /2; − 23 + 23 1

(Eq. 6)

where A and B are 4-hydroxythioanisole and H2O2, respectively.

258 259

Results and Discussion

260 261

Microbial Peroxide Producing Cell (MPPC) operation and H2O2 production

262

We operated two MPPCs with a fixed anode potential of -0.3V (vs. Ag/AgCl) and gas

263

diffusion electrodes optimized for H2O2 production for six months. Bioelectrochemical current

264

began after 5 days and rapidly entered an exponential growth phase. Current density peaked 10

265

days after inoculation at 1.4 mA cm-2 but declined to an average of 0.8 ± 0.3 mA cm-2 for the

266

duration of the experiment. Long-term cell current for one cell is shown in Figure S2. Consistent

267

with previously reported long-term BESs, differences in current maxima before and after


Page 11 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


268

replacing media were apparent after short downtime periods.37 Current density varied during

269

operation due to recurring batch acetate removal experiments, feed interruptions, and

270

maintenance breaks during operation. After startup, average acetate removal was 8.7 ± 2.7 mg

271

cm-2 day-1 (normalized to anode surface area), with a anodic coulombic efficiency of 61 ± 22%

272

(Figure S3). During operation, attached growth on the sides of the reactor was observed, likely

273

due to the growth of microaerobic bacteria that could grow under the hypoxic reactor conditions

274

(

A hybrid approach for selective sulfoxidation via bioelectrochemically

Recommend Documents